Multi-Centre Gastroschisis Interventional Study Across Sub-Saharan Africa

Developing and Implementing an Interventional Bundle to Reduce Mortality From Gastroschisis in Low-Resource Settings

Korle-Bu Teaching Hospital, Accra, Ghana, Komfo Anokye Teaching Hospital, Tamale Teaching Hospital, Tamale, Ghana, University Teaching Hospital, Lusaka, Zambia, Arthur Davison Children's Hospital, Kamuzu Central Hospital, Muhimbili National Hospital, Dar es Salaam, Tanzania.

This study is a multi-centre interventional study at seven tertiary paediatric surgery centres in Ghana, Zambia, Malawi and Tanzania aimed at reducing mortality from gastroschisis.

Background: Gastroschisis has a great disparity in outcomes globally with less than 4% mortality in high-income countries (HICs) and over 90% mortality in many tertiary paediatric surgery centres across sub-Saharan Africa (SSA), including the study centres. Low-technology, cost-effective, evidence-based interventions are available with the potential for improved outcomes in low-resource settings. Aim: To develop, implement and prospectively evaluate an interventional care bundle to reduce mortality from gastroschisis in seven tertiary paediatric surgery centres across SSA. Study Design: A hybrid type 2 effectiveness-implementation study design will be utilised, which focusses equally on the effectiveness of the intervention through analysing clinical outcomes and effectiveness of the implementation through analysis of service delivery and implementation outcomes. Methods: Using current literature an evidence-based, low-technology interventional care bundle has been developed with both pre-hospital and in-hospital components. During 2018 a systematic review, qualitative study and Delphi process will provide further evidence in order to optimise the interventional bundle and implementation strategy for use in low-resource environments. The interventional care bundle has core components, which will remain consistent across the seven sites, and adaptable components, which will be determined through in-country co-development meetings with all relevant members of the multi-disciplinary team. Across the seven study sites it is estimated that 180-289 patients will be included in the study, which has the potential to determine a difference in pre- and post-intervention mortality between 10-15%. Implementation will take place over a 4-week period at each of the study sites between April - December 2019. Pre- and post-interventional data will be collected on clinical, service delivery and implementation outcomes between September 2018 to October 2020 (2-years of patient recruitment at each site with a 30-day post primary intervention follow-up period). The primary clinical outcome will be all-cause, in-hospital mortality. Secondary outcomes include the occurrence of a major complication, length of hospital stay and time to full enteral feeds. Service delivery outcomes include time to hospital and primary intervention and adherence to the pre-hospital and in-hospital management protocols. Implementation outcomes will include acceptability, adoption, appropriateness, feasibility, fidelity, coverage, cost and sustainability. Expected Results: Pre- and post-intervention clinical outcomes will be compared using Chi-squared analysis, unpaired t-test and/ or Mann-Whitney test. Time-series analysis will be undertaken using Statistical Process Control to identify significant trends and shifts in outcome overtime. Multivariate logistic regression analysis will be used to identify both clinical and implementation factors affecting outcome with adjustment for confounders. P<0.05 will be deemed significant. Outcome: This will be the first multi-centre interventional study to the investigator's knowledge aimed at reducing mortality from gastroschisis in low-resource settings. If successful, detailed evaluation of both the clinical and implementation components of the study will allow sustainability in the study sites and further scale-up.

Gastroschisis, Sub-Saharan Africa, Low-resource settings, Low- and middle-income countries, Mortality, Outcomes, Implementation science, Intervention

Pre and post study design. All patients presenting with simple gastroschisis during the pre-intervention phase will receive the current care provided by the study site. All patients with simple gastroschisis presenting to a study site following implementation will receive the interventional care bundle (if consent is provided).

Patients presenting with simple gastroschisis during the pre-intervention phase will receive the current care provided by the study sites (no intervention).

Patients presenting with simple gastroschisis during the post-implementation phase will receive the interventional care bundle if consent for participation is provided.

The interventional care bundle will include pre-hospital and in-hospital components and core and adaptable components. All seven centres will implement the core pre-hospital and in-hospital components. Implementation of the adaptable components will be decided by the local team to optimise the interventional care bundle to the local environment. Pre-hospital core components: 1) covering the bowel in clear plastic, 2) administering intravenous fluids, 3) keeping the neonate warm, 4) transferring to a study centre as soon as possible. This will be implemented as a pre-hospital gastroschisis care protocol to district and secondary level hospitals who refer patients with gastroschisis to the study centres. Pre-hospital adaptable components: 1) the strategy for disseminating the pre-hospital gastroschisis care protocol.

In-hospital core components: 1) use of a standardised care protocol, 2) neonatal resuscitation and ward care including intravenous access, intravenous fluids, maintenance of normothermia, appropriate antibiotics, regular monitoring and infection control, 3) gastroschisis reduction and sutureless closure using a preformed silo and avoidance of neonatal anaesthesia and surgery, 4) early establishment of breastfeeding and an enhanced enteral feeding programme. In-hospital adaptable components: 1) administration of parenteral nutrition for neonates who have survived for 1-week in hospital, 2) maternal involvement in monitoring and basic management, 3) management of neonates with gastroschisis on the neonatal intensive care unit if available.

Mortality whilst in hospital during primary admission, up to 30-days post-intervention or 30-days from admission for those not receiving an intervention and still in hospital.

Major complications are secondary bowel ischaemia, necrosis or perforation requiring resection**, abdominal compartment syndrome*** or need for further unplanned surgical intervention. *Defined as the first bowel intervention the neonate received including silo application or primary closure. **Defined as bowel necrosis or perforation that was not present prior to primary intervention and hence occurred after primary intervention. ***Defined as respiratory insufficiency secondary to compromised tidal volumes, decreased urine output by falling renal perfusion (<1ml/kg/hr) and any other organ dysfunction caused by increased intra-abdominal pressure (based on clinical judgement - the pressure does not have to be measured to fulfil the criteria).

Service delivery outcomes: proportion of patients with clear plastic bowel coverage prior to arrival to the study centre

Service delivery outcomes: proportion of patients with intravenous fluids prior to arrival at hospital.

Service delivery outcomes: position of multi-disciplinary team members involved during the initial resuscitation, primary intervention and defect closure

Service delivery outcomes: proportion of patients receiving parenteral nutrition at some point during their hospital study

Implementation outcome: acceptibility [assessed using the validated 'Acceptability of Intervention Measure' (AIM) Survey]

Acceptability: Perception amongst stakeholders that the new intervention is agreeable. The AIM Survey has 4 questions scored from 1 (completely disagree) to 5 (completely agree). The maximum score is 20 indicating the highest level of acceptability. The lowest score is 4 indicating that the intervention is completely unacceptable.

Survey to be administered at 2 time points: 1) during the 4th week of the implementation phase of the study (1 week) 2) on completion of the study (4 weeks).

Adoption: Intention to apply new intervention. This will be assessed using the validated NoMAD Survey. The NoMAD Survey has questions categorised into 4 themes regarding the intervention: 1) understanding/ opinion, 2) participation, 3) action, 4) monitoring. Each question is scored from 1 (strongly disagree) to 5 (strongly agree). There are 20 questions in total with a maximum score of 100. There are an additional 3 questions regarding the staff members familiarity and use of the intervention scored from 0-10, with 30 being the highest score. Hence, the highest overall score is 130 and lowest 20.

Survey to be administered at 2 time points: 1) during the 1st week of the implementation phase of the study (1 week) 2) on completion of the study (4 weeks).

Implementation outcome: appropriateness [assessed using the validated 'Intervention Appropriateness Measure' (IAM) Survey].

Appropriateness: Perceived relevance of the intervention for the setting and problem.The IAM Survey has 4 questions scored from 1 (completely disagree) to 5 (completely agree). The maximum score is 20 indicating the highest level of intervention appropriateness for the setting. The lowest score is 4 indicating that the intervention is completely inappropriate.

Survey to be administered at 2 time points: 1) during the 4th week of the implementation phase of the study (1 week) 2) on completion of the study (4 weeks).

Implementation outcome: feasibility [assessed using the validated 'Feasibility of Intervention Measure' (FIM) Survey].

Feasibility: Extent to which an intervention can be applied. The FIM Survey has 4 questions scored from 1 (completely disagree) to 5 (completely agree). The maximum score is 20 indicating the highest level of feasibility. The lowest score is 4 indicating that the intervention is completely unfeasible for the setting.

Survey to be administered at 2 time points: 1) during the 4th week of the implementation phase of the study (1 week) 2) on completion of the study (4 weeks).

The proportion of management protocol components completed as intended will be assessed using a checklist at the time of preformed silo application and defect closure.

Total number and position of the MDT members engaged in the qualitative study and/ or Delphi process.

Did an in-person co-development meeting take place at the start of the implementation phase at each study site (yes/ no).

If an in-person co-development meeting took place at the start of the implementation phase at each study site were adaptations made to the interventional bundle accordingly (yes/no).

If an in-person co-development meeting took place at the start of the implementation phase at each study site, the total number and position of MDT members present during the meeting.

Proportion of referral hospitals receiving implementation of the pre-hospital management protocol through in-person education and training.

The proportion of eligible patients who actually receive the intervention will be determined through the data entered into REDCap on all patients presenting with gastroschisis. Neonates with simple gastroschisis included within the study will have all data points completed. Neonates with complex gastroschisis excluded from the study intervention will have baseline data collected on the following: patient demographics, pre-hospital care and outcomes.

The average cost per patient with gastroschisis will be estimated at each of the study centres, pre- and post-intervention. The implementation costs will also be calculated. The number of disability-adjusted life years (DALYs) averted through implementation of the interventional bundle will be calculated. The previously utilised disability-weight of 0.2 will be used for surviving neonates with gastroschisis where 0 is no disability and 1 is death 2. The cost in US$ per DALY averted will be calculated.

Implementation outcome: sustainability [assessed using the NHS Quality Improvement Sustainability Model]

This model quantitatively evaluates process, staff and organisation providing a score for the overall likelihood of sustainability for a newly implemented intervention. The maximum score is 100, with a score of over 55 deemed 'reason for optimism'.

At 3 specific time points in the study: 1) At the time of protocol development (3 months), 2) just prior to implementation of the interventional care bundle (1 week), 3) upon completion of the study (1 week).

Inclusion Criteria: all neonates presenting primarily to the study centre with simple gastroschisis regardless of weight, gestational age or co-morbidities. Exclusion Criteria: all neonates with 'complex gastroschisis' requiring surgical intervention for bowel necrosis, perforation, atresia or other reason.

Wright N, Abantanga F, Amoah M, Appeadu-Mensah W, Bokhary Z, Bvulani B, Davies J, Miti S, Nandi B, Nimako B, Poenaru D, Tabiri S, Yifieyeh A, Ade-Ajayi N, Sevdalis N, Leather A. Developing and implementing an interventional bundle to reduce mortality from gastroschisis in low-resource settings. Wellcome Open Res. 2019 Mar 8;4:46. doi: 10.12688/wellcomeopenres.15113.1. eCollection 2019.